monotone-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Monotone-devel] Re: Bug#404616: Bug#405599: sp_counted_base_gcc_ppc: in


From: Shaun Jackman
Subject: [Monotone-devel] Re: Bug#404616: Bug#405599: sp_counted_base_gcc_ppc: insufficient constraints on inline assembly
Date: Thu, 4 Jan 2007 21:36:46 -0700

package monotone
block 404616 405599
thanks

Thanks for tracking down the root cause, Aaron! I suspected a Boost bug.

Cheers,
Shaun

On 1/4/07, Aaron M. Ucko <address@hidden> wrote:
I think bug #405599 may be responsible for #404616 (which I tried to
x-debbugs-cc):

> Package: libboost-dev
> Version: 1.33.1-9
> Severity: important
> File: /usr/include/boost/detail/sp_counted_base_gcc_ppc.hpp
> Tags: patch
>
> Several functions in sp_counted_base_gcc_{ppc,ia64}.hpp specify their
> read-write *pw argument only as an output operand to their inline
> assembly code, potentially running afoul of gcc's optimizer.  (This
> can definitely be a problem with GCC 4.0.1 on Mac OS X, and I know of
> no reason it wouldn't also affect 4.1.1 on Linux.)
>
> 
http://www.nabble.com/missing-__volatile__-tags-in-sp_counted_base_gcc_xxx.hpp-t1274670.html
>
> has more details, and Boost's upstream developers committed fixes last
> March:
>
> 
http://boost.cvs.sourceforge.net/boost/boost/boost/detail/sp_counted_base_gcc_ia64.hpp?r1=1.2&r2=1.3&view=patch
> 
http://boost.cvs.sourceforge.net/boost/boost/boost/detail/sp_counted_base_gcc_ppc.hpp?r1=1.4&r2=1.5&view=patch
>
> This could conceivably be the root cause of #404616.

--
Aaron M. Ucko, KB1CJC (amu at alum.mit.edu, ucko at debian.org)
Finger address@hidden (NOT a valid e-mail address) for more info.






reply via email to

[Prev in Thread] Current Thread [Next in Thread]